摘 要:学生成绩预测及告警是协助高等院校职能部门人员管理学生学习情况和监测教师教学质量的有效方法。通过提前预测学生考试分数,对预测成绩偏低的学生加强管理,提前重点关注该类学生的学习情况,可以降低学生考试不及格的风险。文章利用矩阵分解算法提取学生历史分数中的关键特征值,并将其用于学生成绩预测。实验结果表明,使用改进的奇异值分解方法预测学生成绩,具有较强的实用性和较好的效果。
关键词:成绩预测;矩阵分解;奇异值分解;学业预警
中图分类号:G434;TP391 文献标识码:A 文章编号:2096-4706(2020)04-0142-04
Research on College Students’Performance Prediction Based on Matrix Decomposition
XUE Mengting
(Guangdong University of Foreign Studies,Guangzhou 510420,China)
Abstract:The prediction and warning of students’performance is an effective method to assist teachers of the functional departments of colleges to manage students’learning and monitor teachers’teaching quality. By predicting students’examination scores in advance,strengthening the management of students with low predicted scores,and focusing on the learning of such students in advance can reduce them the risk of failing the exam. In this paper,we use the matrix factorization algorithms to extract the key eigenvalues from the student’s historical scores and use them to predict student performance. The experimental results show that the improved SVD method is practical and effective in predicting students’performance.
Keywords:performance prediction;matrix decomposition;singular value decomposition;academic warning
参考文献:
[1] 王运武,于长虹 . 智慧校园——实现智慧教育的必由之路 [M]. 北京:电子工业出版社,2016.
[2] 陈子健,朱晓亮 . 基于教育数据挖掘的在线学习者学业成绩预测建模研究 [J]. 中国电化教育,2017(12):75-81+89.
[3] 肖巍,倪传斌,李锐 . 国外基于数据挖掘的学习预警研究:回顾与展望 [J]. 中国远程教育,2018(2):70-78.
[4] 刘艳杰,李霞 . 基于贝叶斯网络的学生成绩预测 [J]. 山东理工大学学报(自然科学版),2019,33(5):75-78.
[5] 张莉,卢星凝,陆从林,等 . 支持向量机在高考成绩预测分析中的应用 [J]. 中国科学技术大学学报,2017,47(1):1-9.
[6] FUNK S. Netflix Update:Try This At Home [EB/OL].(2006-12-11).https://sifter.org/~simon/journal/20061211.html.
[7] BELL R M,KOREN Y,VOLINSKY C. Research Track Paper Modeling Relationships at Multiple Scales to Improve Accuracy of Large Recommender Systems [C]// Acm Sigkdd International Conference on Knowledge Discovery & Data Mining. ACM,2007:95-104.
[8] KOREN Y. Factorization meets the neighborhood:A multifaceted collaborative filtering model [C]// Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,Las Vegas,Nevada,USA,August 24-27,2008. ACM,2008:426-434. [9] LEE D D,SEUNG H S. Learning the parts of objects by non negative matrix factorization [J].Nature,1999,401(6755):788-791.
作者简介:薛梦婷(1992.07-),女,汉族,江苏南京人,助理实验师,硕士研究生,研究方向:数据挖掘、中文情感分析、教育信息化。